Connect
← Return to Patient Experience and AI
Newsfeed PostNewsfeed Post in blog Ehlers-Danlos Syndrome
Patient Experience and AI 17 hours ago | By MeekaC (@meekaclayton)Comment receiving replies
I have tried without success to have Claude create a verbatim, analyzable repository of my medical records. I am sure part of the problem is that I am untrained in how to do this, but the issue for me is that the software repeatedly inserted its opinions and analysis in my records, corrupting them. Do you also face this problem or have suggestions for avoiding it?
Replies to "I have tried without success to have Claude create a verbatim, analyzable repository of my medical..."
Connect
@cdammen, I encourage you to read this article from Mayo Clinic about how to use AI safely for health questions.
- Can you trust AI for health advice? https://connect.mayoclinic.org/blog/about-connect/newsfeed-post/can-you-trust-ai-for-health-advice/
I underline the bullet point:
"Protect your privacy. Don't share personal information, like your full name, date of birth, address, medical records or insurance details. Even health details should be shared cautiously, especially on public or free platforms."
When you share your medical records with an AI tool, like Claude, ChatGPT, Perplexity, etc., it is no longer private.
The AI tools that Mayo Clinic differ from the publicly available versions and have strong privacy and security built in to a closed system.
I see you've discovered the sychophant nature of AI agents. This means that the tools are built to tailor their responses to validate or agree with human preferences rather than providing objective facts or critical thinking. There are situations where that can be misleading or even cause harm.